RWMAC's Advice to Ministers on the Radioactive Waste Implications of Reprocessing

4. ASSEMBLY AND INTERPRETATION OF DATA

4.1 Sources of data

For the moment, only British Nuclear Fuels plc (BNFL) undertake the reprocessing of spent nuclear fuel within the UK. Previously reprocessing has also been undertaken at the United Kingdom Atomic Energy Authority (UKAEA) establishment at Dounreay. However, an end to commercial reprocessing at Dounreay was announced by the UK Government in June 1998 and the site's main reprocessing facility is currently out of operation pending a decision on whether or not to replace the fuel dissolver. A recent consultation document issued by the Department of Trade and industry on the basis of material prepared by UKAEA¹² suggested that there was only a limited amount of fast reactor fuel that might be subject to future reprocessing, although other possibilities were mooted. The handling of a limited amount of spent fuel from the naval submarine programme is also not covered. The amounts of spent fuel from these other sources are extremely small in relation to the amounts likely to be reprocessed at Sellafield. For this reason, RWMAC restricted its current study to Sellafield operations.

Inevitably, RWMAC had to request the majority of the information and data required for its work from BNFL. Some of this information that has not previously been placed in the public domain. The Committee undertook all the checks that it felt were reasonable in respect of the data and information provided by BNFL. It questioned derivation, followed calculations through insofar as was reasonably possible, reviewed radiation dose estimation models and commented on any obvious differences from other published material. Prior to finalisation of this report it also asked BNFL to vouch for the factual accuracy of its content. However, in the final analysis, the Committee must acknowledge that it could not review each and every aspect of the data's production within the resources available to it, and thus had to rely on the company's input and expertise.

4.2 Scenarios considered

RWMAC agreed with BNFL early on in its work a number of possible reprocessing scenarios against which to carry out analyses. These were required to be representative of a range encompassing all the possible future options. The reference date against which the outcome of the various scenarios were analysed was 1 April 2000.

The scenarios considered are set out and described in Table 1. In choosing them, it was necessary to distinguish between the different possibilities for Magnox and THORP reprocessing (see Annex 1), because of the different nature and aims of the Magnox reprocessing programme on the one hand and the THORP programme for the reprocessing of AGR and PWR fuel on the other.

It should be noted that the figures quoted for the weights of spent fuel in Table 1 and the remainder of this report follow the nuclear industryconvention ofbeing given in terms of tonnes of heavy metal, abbreviated to tHM (t being the abbreviation for tonnes and HM being short for heavy metal). This is the content of uranium and plutonium metal present in the materials under consideration.

Table 1

Reprocessing scenarios considered

The Table 1 scenarios aimed to extend, for both Magnox and THORP, from the earliest practicable end to reprocessing, through BNFL's business plan, or Reference scenario, existing at the time the RWMAC study commenced, to the company's most optimistic future business assumption. The Reference scenarios originally chosen were essentially in line with those used to derive forecasts of waste arisings for the 1998 UK Radioactive Waste Inventory¹³.

For the purpose of estimating total radioactive materials arisings, three combined Magnox and THORP scenarios (again see Table 1) were considered so as to be able to assess the complete range of possible future outcomes. These were: a Combined Extended scenario (C1), a Combined Reference scenario (C2), and a Combined Early Termination scenario (C3). The analysis of these three combined scenarios ultimately provided the main basis for RWMAC assessments and conclusions set out in this report.

A key issue in deciding which Magnox scenarios should be included in the Combined Early Termination scenario concerned the need for reprocessing of Magnox spent fuel. When stored under water, as is the case at all UK Magnox sites other than Wylfa, Magnox spent fuel undergoes corrosion, initially of the magnesium alloy cladding and subsequently of the uranium metal within. Once corrosion has started, reprocessing is inevitable to transform it into materials which are ultimately suitable for storage and potentially also, in due course, for underground disposal.

The alternative that has been mooted is a strategy of dry storage of the Magnox spent fuel, which might remove the need for its reprocessing. Wylfa has a dry air store but the stored fuel is eventually dispatched for wet storage at Sellafield prior to reprocessing. During the course of its work, RWMAC asked BNFL to supply a view of the dry storage issue, commenting specifically on the Wylfa situation. The written statement that they supplied, together with a RWMAC commentary on it, is contained in Annex 4 to this report. The view that RWMAC arrived at, as a result of this analysis, may be summarised as follows.

Both uranium metal contained in spent Magnox fuel and its magnesium alloy cladding are highly reactive materials. Even if uranium metal spent fuel were dry stored, it would need to be in a carefully controlled environment and would not be line with the concept of passivity (see section 6.1). If a policy of deep disposal is ultimately pursued, the fuel would then also subsequently have to be suitably conditioned, given that it would still be reactive in a geological environment. Currently, no industrial-scale process exists to do this, other than reprocessing itself. It is therefore rational to continue to reprocess the spent fuel while the B205 facility exists. It also implies that if Magnox reprocessing in B205 had to stop, the Magnox reactors themselves would also have to close.

If an alternative existed to reprocessing as a means of conditioning dry-stored spent fuel for eventual deep disposal, then the above chain of logic would fail. A substantial proportion of the spent fuel produced and wet-stored would inevitably have to be reprocessed, because of likely dry store development timescales. However, RWMAC is not convinced that dry storage for the remainder need then be ruled out, for an interim period, on technical regulatory, or economic grounds. In such circumstances, the cessation of Magnox reprocessing would not necessarily force closure of the Magnox stations. However, planning on the basis of such a hypothetical eventuality would clearly be most unwise and, in the absence of any real prospect of any alternative conditioning strategy emerging, reprocessing of Magnox spent fuel must continue. As will be made clear later in this report (see sections 6.3 and 7.5), if B205 reprocessing throughputs cannot be maintained at appropriate levels, it is liable to put the UK Government's proposed OSPAR objectives at risk.

Thus, RWMAC accepted that securing an alternative to the reprocessing of spent Magnox fuel is not a practicable possibility at the present time. This is why the Magnox 3 scenario, rather than the Magnox 4 scenario is included in the Combined Early Termination scenario. That is, no more fuel would be loaded into the Magnox reactors, but all existing fuel - taken to be 7,700 tHM in the reactor or storage ponds - is reprocessed. However, given its potential sensitivity, it is suggested that the issue of the need to reprocess Magnox spent fuel is one that should be addressed in the Government's forthcoming solid radioactive waste management policy consultation.

THORP reprocessing is different, because the uranium oxide contained in AGR and LWR reactor fuel is much more stable than uranium metal, being found in nature as uranium ores. Thus THORP reprocessing scenarios all assume that the UK's AGR reactors will continue to operate to the scenario set out in the 1998 UK Radioactive Waste Inventory¹³, irrespective of how much of the spent fuel is reprocessed, with ongoing storage being the option adopted for the remainder. THORP reprocessing is also based on the reprocessing of overseas fuel, the precise amount of which depends on overseas orders won.

In the event, BNFL's Magnox station lifetime strategy changed during the course of the RWMAC work. On 23 May 2000, the company announced the revised lifetime strategy set out in Table 2. The differences in anticipated reactor lifetime from those incorporated in the Magnox 2 Reference Scenario are shown. The overall effect of the revised plan was to extend the expected operating lives of the majority of the Magnox reactors.

Table 2

Magnox station lifetime figures

In its 23 May 2000 statement, BNFL noted that continuing to run Oldbury and Wylfa to these revised dates depended upon the development and use of Magrox fuel. Magrox is intended to be a ceramic uranium oxide, as opposed to a uranium metal, fuel enclosed in a stainless steel, as opposed to a magnesium alloy, cladding that can be used in Magnox reactors. A decision on the use of Magrox fuel will be taken in around 2003 following the results of reactor trials. Assuming a decision was taken to proceed and the necessary regulatory approvals could be secured, BNFL anticipate some five years to make the full transition of the reactor cores to Magrox fuel. Oldbury and Wylfa will also need to undergo a Periodic Safety Review by the Nuclear Installations Inspectorate in order to achieve extended operation.

In this context, it should be noted that the Oldbury and Wylfa stations are the more modern, larger capacity concrete (as opposed to steel) pressure vessel reactors which, between them, are responsible for more than 50 per cent of Magnox electricity generation. If Magrox fuel were not practicable, then closure of these stations would occur around 2010. BNFL consider that the lives quoted in Table 2 are the longest durations the Magnox reactors will operate. Market conditions and any future technical problems could result in premature closure before these dates. BNFL stated the belief that this plan would enable the Magnox reprocessing plant (B205) to be closed around 2012.

What the revised BNFL plan implies for the RWMAC study is that Magnox Reference Scenario analysis will underestimate the revised BNFL business case Magnox materials arisings. The revised business case results in total forecast Magnox spent fuel arisings of 1,500 tHM greater than the Table 1 Magnox 2 Reference scenario. It was impracticable within the time available for this study to revise all the analyses undertaken, which in any case remain sufficient for the illustrative purposes of this study.